Patients with non-Hodgkin lymphoma (NHL) or chronic lymphocytic leukemia (CLL) respond well to chimeric antigen receptor (CAR)-redirected T cells specific for the CD19 antigen and encoding costimulatory endodomains. Despite encouraging recent reports of therapeutic efficacy, this approach does not distinguish between normal and malignant B cells, setting the stage for profound B cell depletion, hypogammaglobulinemia and perhaps other delayed toxic effects due to the persistence of activated T cells. To circumvent this obstacle without substantial loss of antitumor potency, we are testing in a phase 1 clinical trial (Aim 1) of Project 2, CAR-modified T cells that recognize the Ig kappa light chain of malignant B cells The central hypothesis is that such therapy will eradicate kappa-positive NHL and CLL cells while sparing normal B cells that express the nontargeted light chain, thus preserving B cell function at a critical time in the patient's clinical course. To reduce the likelihood of suboptimal cell killing due to varying levels or a complete loss of tumor antigen expression, we have chosen CD23 - which is independently expressed by CLL cells - as a second target antigen. This modification could boost the tumor cell kill achieved with T cells that recognize kappa light chain, a prediction we intend to test in a preclinical model (Aim 2). In an additional model we plan to add selected immunomodulatory drugs to the T cell regimen to subvert the immune-inhibitory tumor micro-environment, overcoming tumor-associated neoangiogenesis, Treg cell recruitment and inhibitory immune.cytokines (Aim 3), all of which continue to impede the development of effective T cell immunotherapy. Better control of tumor cell evasion tactics is expected to enhance antitumor effects beyond those typically seen with use of CAR-modified T cells alone. The information gained from the clinical evaluation in Aim 1, together with the preclinical studies in Aims 2 and 3, should inform the design of